Commit f2448c93 authored by Martin Hansson's avatar Martin Hansson

Bug#44821: select distinct on partitioned table returns wrong results

Range analysis did not request sorted output from the storage engine,
which cause partitioned handlers to process one partition at a time
while reading key prefixes in ascending order, causing values to be 
missed. Fixed by always requesting sorted order during range analysis.
This fix is introduced in 6.0 by the fix for bug no 41136.

mysql-test/r/group_min_max.result:
  Bug#44821: Test result.
mysql-test/t/group_min_max.test:
  Bug#44821: Test case
sql/opt_range.cc:
  Bug#44821: Fix.
parent ed7f0f30
...@@ -2463,3 +2463,61 @@ c ...@@ -2463,3 +2463,61 @@ c
2 2
DROP TABLE t1; DROP TABLE t1;
End of 5.0 tests End of 5.0 tests
CREATE TABLE t1 ( a INT, b INT, c INT, KEY bc(b, c) )
PARTITION BY KEY (a, b) PARTITIONS 3
;
INSERT INTO t1 VALUES
(17, 1, -8),
(3, 1, -7),
(23, 1, -6),
(22, 1, -5),
(11, 1, -4),
(21, 1, -3),
(19, 1, -2),
(30, 1, -1),
(20, 1, 1),
(16, 1, 2),
(18, 1, 3),
(9, 1, 4),
(15, 1, 5),
(28, 1, 6),
(29, 1, 7),
(25, 1, 8),
(10, 1, 9),
(13, 1, 10),
(27, 1, 11),
(24, 1, 12),
(12, 1, 13),
(26, 1, 14),
(14, 1, 15)
;
SELECT b, c FROM t1 WHERE b = 1 GROUP BY b, c;
b c
1 -8
1 -7
1 -6
1 -5
1 -4
1 -3
1 -2
1 -1
1 1
1 2
1 3
1 4
1 5
1 6
1 7
1 8
1 9
1 10
1 11
1 12
1 13
1 14
1 15
EXPLAIN
SELECT b, c FROM t1 WHERE b = 1 GROUP BY b, c;
id select_type table type possible_keys key key_len ref rows Extra
1 SIMPLE t1 range bc bc 10 NULL 7 Using where; Using index for group-by
DROP TABLE t1;
...@@ -983,3 +983,43 @@ SELECT DISTINCT c FROM t1 WHERE d=4; ...@@ -983,3 +983,43 @@ SELECT DISTINCT c FROM t1 WHERE d=4;
DROP TABLE t1; DROP TABLE t1;
--echo End of 5.0 tests --echo End of 5.0 tests
#
# Bug#44821: select distinct on partitioned table returns wrong results
#
CREATE TABLE t1 ( a INT, b INT, c INT, KEY bc(b, c) )
PARTITION BY KEY (a, b) PARTITIONS 3
;
INSERT INTO t1 VALUES
(17, 1, -8),
(3, 1, -7),
(23, 1, -6),
(22, 1, -5),
(11, 1, -4),
(21, 1, -3),
(19, 1, -2),
(30, 1, -1),
(20, 1, 1),
(16, 1, 2),
(18, 1, 3),
(9, 1, 4),
(15, 1, 5),
(28, 1, 6),
(29, 1, 7),
(25, 1, 8),
(10, 1, 9),
(13, 1, 10),
(27, 1, 11),
(24, 1, 12),
(12, 1, 13),
(26, 1, 14),
(14, 1, 15)
;
SELECT b, c FROM t1 WHERE b = 1 GROUP BY b, c;
EXPLAIN
SELECT b, c FROM t1 WHERE b = 1 GROUP BY b, c;
DROP TABLE t1;
...@@ -8556,7 +8556,7 @@ int QUICK_RANGE_SELECT::get_next_prefix(uint prefix_length, ...@@ -8556,7 +8556,7 @@ int QUICK_RANGE_SELECT::get_next_prefix(uint prefix_length,
result= file->read_range_first(last_range->min_keypart_map ? &start_key : 0, result= file->read_range_first(last_range->min_keypart_map ? &start_key : 0,
last_range->max_keypart_map ? &end_key : 0, last_range->max_keypart_map ? &end_key : 0,
test(last_range->flag & EQ_RANGE), test(last_range->flag & EQ_RANGE),
sorted); TRUE);
if (last_range->flag == (UNIQUE_RANGE | EQ_RANGE)) if (last_range->flag == (UNIQUE_RANGE | EQ_RANGE))
last_range= 0; // Stop searching last_range= 0; // Stop searching
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment